
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

%matplotlib inline

pd.options.mode.chained_assignment = None #disabling the "SettingWithCopyWarning".


df_npp = pd.read_csv("./Datasets/Nuclear Power Plants.csv")
df_npp


# I'm creating a copy because I will need it later.

df_npp_0 = df_npp.copy()


df_npp["Facility Name | Nom de l'installation"].unique()

array(['Gentilly-2', 'Point Lepreau Generating Station',
       'Bruce Power - A', 'Bruce Power - B', 'Bruce Power Site',
       'Darlington Nuclear', 'Pickering Nuclear - A & B',
       'Pickering Nuclear - A', 'Pickering Nuclear - B',
       'Bruce Power Site '], dtype=object)


# Renaming columns to English only:

df_npp.rename(columns={'Year | Année': 'Year', 'NPRI ID | ID INRP': 'NPRI ID','Company Name | Raison Sociale':'Company Name',"Facility Name | Nom de l'installation":'Facility Name', 'City | Ville':'City', 'CSD | SDR':'CSD','CA or CMA | AR ou RMR':'CA or CMA', 'Economic Region | Région économique':'Economic Region','Province | Province':'Province', 'Latitude | Latitude':'Latitude', 'Longitude | Longitude':'Longitude','Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)','Units | Unités':'Units', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge','Footnotes | Notes de bas de page':'Footnotes'}, inplace = True)
df_npp.head()


# Stack Emission column first:

df_npp_miss_stack = df_npp[df_npp['Stack Emissions'].isin(['LD', 'NRM | NRS'])]
df_npp_miss_stack[['Year','Facility Name', 'Substance Name (English)', 'Stack Emissions']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


# Direct Discharge column next:

df_npp_miss_discharge = df_npp[df_npp['Direct Discharge'].isin(['LD', 'NRM | NRS'])]
df_npp_miss_discharge[['Year','Facility Name', 'Substance Name (English)', 'Direct Discharge']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


# I noticed one value is "0":

df_npp[(df_npp['Stack Emissions'] == '0.00E+00') | (df_npp['Direct Discharge'] == '0.00E+00')]


# Combining Bruce Power A & B and Pickering Nuclear A & B for a geographic reference table, so I can delete them from the dataframe I will use for plotting:

df_npp_geography = df_npp[['Facility Name', 'NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude']]
df_npp_geography['Facility Name'].replace('Bruce Power - A', 'Bruce Power Site', inplace=True)
df_npp_geography['Facility Name'].replace('Bruce Power - B', 'Bruce Power Site', inplace=True)
df_npp_geography['Facility Name'].replace('Bruce Power Site ','Bruce Power Site',inplace=True)
df_npp_geography['Facility Name'].replace('Pickering Nuclear - A', 'Pickering Nuclear - A & B', inplace=True)
df_npp_geography['Facility Name'].replace('Pickering Nuclear - B', 'Pickering Nuclear - A & B', inplace=True)
df_npp_geography


# Cleaning the geography dataframe:

df_npp_geography.drop_duplicates(inplace=True)
df_npp_geography = df_npp_geography.reset_index(drop=True)
df_npp_geography


df_npp_geography.at[5, 'NPRI ID'] = 3161
df_npp_geography.drop_duplicates(inplace=True)
df_npp_geography = df_npp_geography.reset_index(drop=True)
df_npp_geography


# Cleaning the LD (Level of detection) & NRM (Not required to monitor) values so I can convert the columns into numeric for later plotting:

df_npp['Stack Emissions'].replace('LD', 0, inplace=True)
df_npp['Stack Emissions'].replace('NRM | NRS', 0, inplace=True)
df_npp['Direct Discharge'].replace('LD', 0, inplace=True)
df_npp['Direct Discharge'].replace('NRM | NRS', 0, inplace=True)


# At line 28 there is a problem with a figure. It says 1,2e12 instead of 1.2e12. I will correct it:

df_npp.at[28, 'Stack Emissions'] = 1.2e12


# Converted columns to numeric for plotting:

df_npp['Stack Emissions'] = pd.to_numeric(df_npp['Stack Emissions'])
df_npp['Direct Discharge'] = pd.to_numeric(df_npp['Direct Discharge'])


# Combining Pickering Nuclear A & Pickering Nuclear B so I can plot every year (they are reported combined after 208). Also correcting the name of one data point of Bruce Power:

df_npp['Facility Name'].replace('Bruce Power Site ','Bruce Power Site',inplace=True)
df_npp['Facility Name'].replace('Pickering Nuclear - A', 'Pickering Nuclear - A & B', inplace=True)
df_npp['Facility Name'].replace('Pickering Nuclear - B', 'Pickering Nuclear - A & B', inplace=True)
df_npp['Facility Name'].unique()

array(['Gentilly-2', 'Point Lepreau Generating Station',
       'Bruce Power - A', 'Bruce Power - B', 'Bruce Power Site',
       'Darlington Nuclear', 'Pickering Nuclear - A & B'], dtype=object)


df_npp.drop(columns=['NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude', 'Substance Name (French) | Nom de substance (Français)', '<', '<.1', 'Footnotes'], inplace=True)
df_npp.head()


# Aggregating Pickering A/B to be able to have a same value accross the years (because in recent years they only report the 'combined' site):

df_npp = df_npp.groupby(['Year', 'Facility Name', 'Substance Name (English)', 'Units'],as_index=False).agg({'Stack Emissions': 'sum', 'Direct Discharge': 'sum'})
df_npp.head()


# I'm saving the clean dataframe to do a dashboard in Tableau.

df_npp.to_csv(".\Datasets\df_npp.csv", index=True, header=True)


df_npp['Substance Name (English)'].unique()

array(['Carbon-14', 'Iodine-131', 'Particulate (gross beta/gamma)',
       'Particulate gross alpha', 'Total noble gases', 'Tritium (HTO)',
       'Estimated public dose (see footnote)', 'Elemental Tritium (HT)'],
      dtype=object)


df_npp_epd = df_npp[df_npp['Substance Name (English)'] == 'Estimated public dose (see footnote)']
df_npp_epd.head()


# Estimated public dose is calculated incorporating all major release pathways (emissions and discharges)

plt.figure(figsize=(16,6))

year = df_npp_epd['Year'].unique()

for facility in df_npp_epd['Facility Name'].unique():
    plt.plot(year, df_npp_epd[df_npp_epd['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Estimated public dose [mSv/a]', size=12)
plt.legend(df_npp_epd['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_npp_tritium = df_npp[df_npp['Substance Name (English)'] == 'Tritium (HTO)']
df_npp_tritium.head()


plt.figure(figsize=(16,6))

year = df_npp_tritium['Year'].unique()

for facility in df_npp_tritium['Facility Name'].unique():
    plt.plot(year, df_npp_tritium[df_npp_tritium['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Tritium (HTO) - Stack Emissions [Bq]', size=12)
plt.legend(df_npp_tritium['Facility Name'].unique(), loc='lower right')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_npp_tritium['Year'].unique()

for facility in df_npp_tritium['Facility Name'].unique():
    plt.plot(year, df_npp_tritium[df_npp_tritium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Tritium (HTO) - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_tritium['Facility Name'].unique(), loc='upper center')
plt.grid()

plt.show()


df_npp_carbon = df_npp[df_npp['Substance Name (English)'] == 'Carbon-14']
df_npp_carbon.head()


plt.figure(figsize=(16,6))

year = df_npp_carbon['Year'].unique()

for facility in df_npp_carbon['Facility Name'].unique():
    plt.plot(year, df_npp_carbon[df_npp_carbon['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Carbon-14 - Stack Emissions [Bq]', size=12)
plt.legend(df_npp_carbon['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_npp_carbon['Year'].unique()

for facility in df_npp_carbon['Facility Name'].unique():
    plt.plot(year, df_npp_carbon[df_npp_carbon['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Carbon-14 - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_carbon['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


# I will plot without it, to see the rest without it:

df_npp_carbon2 = df_npp_carbon[df_npp_carbon['Facility Name'] != 'Gentilly-2']

plt.figure(figsize=(16,6))

year = df_npp_carbon2['Year'].unique()

for facility in df_npp_carbon2['Facility Name'].unique():
    plt.plot(year, df_npp_carbon2[df_npp_carbon2['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Carbon-14 - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_carbon2['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_npp_iodine = df_npp[df_npp['Substance Name (English)'] == 'Iodine-131']
df_npp_iodine.head()


plt.figure(figsize=(16,6))

year = df_npp_iodine['Year'].unique()

for facility in df_npp_iodine['Facility Name'].unique():
    plt.plot(year, df_npp_iodine[df_npp_iodine['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Iodine-131 - Stack Emissions [Bq]', size=12)
plt.legend(df_npp_iodine['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_npp_iodine['Year'].unique()

for facility in df_npp_iodine['Facility Name'].unique():
    plt.plot(year, df_npp_iodine[df_npp_iodine['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Iodine-131 - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_iodine['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_npp_beta = df_npp[df_npp['Substance Name (English)'] == 'Particulate (gross beta/gamma)']
df_npp_beta.head()


plt.figure(figsize=(16,6))

year = df_npp_beta['Year'].unique()

for facility in df_npp_beta['Facility Name'].unique():
    plt.plot(year, df_npp_beta[df_npp_beta['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate (gross beta/gamma) - Stack Emissions [Bq]', size=12)
plt.legend(df_npp_beta['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_npp_beta['Year'].unique()

for facility in df_npp_beta['Facility Name'].unique():
    plt.plot(year, df_npp_beta[df_npp_beta['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate (gross beta/gamma) - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_beta['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


# I'll plot without Pickering to have a better look at the others:

df_npp_beta2 = df_npp_beta[df_npp_beta['Facility Name'] != 'Pickering Nuclear - A & B']

plt.figure(figsize=(16,6))

year = df_npp_beta2['Year'].unique()

for facility in df_npp_beta2['Facility Name'].unique():
    plt.plot(year, df_npp_beta2[df_npp_beta2['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate (gross beta/gamma) - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_beta2['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


# I'll plot without Darlington now:

df_npp_beta3 = df_npp_beta2[df_npp_beta2['Facility Name'] != 'Darlington Nuclear']

plt.figure(figsize=(16,6))

year = df_npp_beta3['Year'].unique()

for facility in df_npp_beta3['Facility Name'].unique():
    plt.plot(year, df_npp_beta3[df_npp_beta3['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate (gross beta/gamma) - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_beta3['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_npp_alpha = df_npp[df_npp['Substance Name (English)'] == 'Particulate gross alpha']
df_npp_alpha.head()


plt.figure(figsize=(16,6))

year = df_npp_alpha['Year'].unique()

for facility in df_npp_alpha['Facility Name'].unique():
    plt.plot(year, df_npp_alpha[df_npp_alpha['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross alpha - Stack Emissions [Bq]', size=12)
plt.legend(df_npp_alpha['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_npp_alpha['Year'].unique()

for facility in df_npp_alpha['Facility Name'].unique():
    plt.plot(year, df_npp_alpha[df_npp_alpha['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross alpha - Direct Discharge [Bq]', size=12)
plt.legend(df_npp_alpha['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_npp_ht = df_npp[df_npp['Substance Name (English)'] == 'Elemental Tritium (HT)']
df_npp_ht.head()


plt.figure(figsize=(16,6))

year = df_npp_ht['Year'].unique()

for facility in df_npp_ht['Facility Name'].unique():
    plt.plot(year, df_npp_ht[df_npp_ht['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Elemental Tritium (HT) - Stack Emissions [Bq]', size=12)
plt.legend(df_npp_ht['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_npp_noble = df_npp[df_npp['Substance Name (English)'] == 'Total noble gases']
df_npp_noble.head()


plt.figure(figsize=(16,6))

year = df_npp_noble['Year'].unique()

for facility in df_npp_noble['Facility Name'].unique():
    plt.plot(year, df_npp_noble[df_npp_noble['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Total noble gases - Stack Emissions [Bq-MeV]', size=12)
plt.legend(df_npp_noble['Facility Name'].unique(), loc='upper center')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_npp_noble['Year'].unique()

for facility in df_npp_noble['Facility Name'].unique():
    plt.plot(year, df_npp_noble[df_npp_noble['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Total noble gases - Direct Discharge [Bq-MeV]', size=12)
plt.legend(df_npp_noble['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


facilities = df_npp['Facility Name'].unique()

for f in facilities:
    df = df_npp[df_npp['Facility Name'] == f]
    print(f,'\n')
    subs = df['Substance Name (English)'].unique()
    for s in subs:
        df2 = df[df['Substance Name (English)'] == s]
        fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(10,8))
        fig.subplots_adjust(hspace=0.5)
    
        ax1.plot(df2['Year'], df2['Stack Emissions'], color='green')
        ax1.set_title(s + ' - Stack Emissions', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax1.set_xlabel('Year', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax1.grid()
    
        ax2.plot(df2['Year'], df2['Direct Discharge'], color='red')
        ax2.set_title(s + ' - Direct Discharge', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax2.set_xlabel('Year', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax2.grid()
    
        plt.show()

Bruce Power - A

Bruce Power - B

Bruce Power Site

Darlington Nuclear

Gentilly-2

Pickering Nuclear - A & B

Point Lepreau Generating Station


df_npp_2020 = pd.read_csv("./Datasets/2020/Nuclear Power Plants.csv")
df_npp_2020.head()


# It doesn't have the same columns, so I will keep only the essentials:

df_npp_2020 = df_npp_2020[['Year | Annee', "Facility Name | Nom de l'installation", 'Substance Name (English) | Nom de substance (Anglais)', 'Stack Emissions | Emissions de cheminees',
       'Direct Discharge | Evacuations directes']]
df_npp_2020.rename(columns={'Year | Annee': 'Year', "Facility Name | Nom de l'installation":'Facility Name', 'Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)', 'Stack Emissions | Emissions de cheminees':'Stack Emissions','Direct Discharge | Evacuations directes':'Direct Discharge'}, inplace = True)
df_npp_2020.head()


# I will go back to the original copy of the dataframe & keep only the essentials to compare:

df_npp_2021 = df_npp_0[['Year | Année', "Facility Name | Nom de l'installation", 'Substance Name (English) | Nom de substance (Anglais)', 'Stack Emissions | Émissions de cheminées',
       'Direct Discharge | Évacuations directes']]
df_npp_2021.rename(columns={'Year | Année': 'Year', "Facility Name | Nom de l'installation":'Facility Name', 'Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge'}, inplace = True)
df_npp_2021.head()


# Now that they have the same columns, I will remove 2021 from the new dataframe and compare the remaining with 2020.

df_npp_2021 = df_npp_2021[df_npp_2021['Year'] != 2021].reset_index(drop = True)
df_npp_2021.head()


# I will concatenate both dataframes & keep the not duplicates to see the differences. This produces a df with 2021's values first, and 2020's values after:

df = pd.concat([df_npp_2021,df_npp_2020]).drop_duplicates(keep=False)
df

# 61 changes total. Some changes are LD to NRM . And others are numerical changes.


# Looking for LD (2020) to NRM (2021) changes first.

df[(df['Stack Emissions'].isin(['NRM | NRS', 'LD'])) & (df['Direct Discharge'] == 'NRM | NRS')] 

# I checked with 'Direct Discharge = NRM/LD', but there wasn't any change there. 
# The second condition is to filter out 2 results that are numerical changes in values of 'Direct Discharge', but same 'Stack Emissions'.
# Finally, it's important to see the Index number 63. I will address that below.


# What's going on with index 63 that it didn't appear twice in the table above (for 2021 & then for 2020):

df.loc[[63]]


# Actual changes in: Index Number: 364, 417, 418, 371, 280, 281, 229, 51, 287, 429, 330, 60, 338. (I filtered out the 'LD' & 'NRM' to get this list, 13 total)

df.loc[[364, 417, 418, 371, 280, 281, 229, 51, 287, 429, 330, 60, 338]] # First value: 2021, Second Value: 2020.


df_npf = pd.read_csv("./Datasets/Nuclear Processing Facilities.csv")
df_npf


# I'm creating a copy because I will need it later.

df_npf_0 = df_npf.copy()


df_npf["Facility Name | Nom de l'installation"].unique()

array(['Blind River Refinery', 'Port Hope Conversion Facility',
       'Cameco Fuel Manufacturing', 'BWXT - Toronto',
       'BWXT - Peterborough', 'SRBT', 'Nordion - Ottawa'], dtype=object)


# Renaming columns to English only:

df_npf.rename(columns={'Year | Année': 'Year', 'NPRI ID | ID INRP': 'NPRI ID','Company Name | Raison Sociale':'Company Name',"Facility Name | Nom de l'installation":'Facility Name', 'City | Ville':'City', 'CSD | SDR':'CSD','CA or CMA | AR ou RMR':'CA or CMA', 'Economic Region | Région économique':'Economic Region','Province | Province':'Province', 'Latitude | Latitude':'Latitude', 'Longitude | Longitude':'Longitude','Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)','Units | Unités':'Units', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge','Footnotes | Notes de bas de page':'Footnotes'}, inplace = True)
df_npf.head()


# Stack Emission column first:

df_npf_miss_stack = df_npf[df_npf['Stack Emissions'] == 'NRM | NRS']
df_npf_miss_stack[['Year','Facility Name', 'Substance Name (English)', 'Stack Emissions']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


# Direct Discharge column next:

df_npf_miss_discharge = df_npf[df_npf['Direct Discharge'] == 'NRM | NRS']
df_npf_miss_discharge[['Year','Facility Name', 'Substance Name (English)', 'Direct Discharge']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


# I noticed some values are "0" or "0.00E+00".

df_npf[df_npf['Stack Emissions'].isin(['0', '0.00E+00'])][['Year','Facility Name', 'Substance Name (English)', 'Stack Emissions', 'Direct Discharge']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


df_npf_geography = df_npf[['Facility Name', 'NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude']]
df_npf_geography


# Cleaning the geography dataframe:

df_npf_geography.drop_duplicates(inplace=True)
df_npf_geography = df_npf_geography.reset_index(drop=True)
df_npf_geography


# Cleaning the NRM (Not required to monitor) values so I can convert the columns into numeric for later plotting:

df_npf['Stack Emissions'].replace('NRM | NRS', 0, inplace=True)
df_npf['Direct Discharge'].replace('NRM | NRS', 0, inplace=True)


# Converted columns to numeric for plotting:

df_npf['Stack Emissions'] = pd.to_numeric(df_npf['Stack Emissions'])
df_npf['Direct Discharge'] = pd.to_numeric(df_npf['Direct Discharge'])


# I noticed in the footnotes of Estimated Public Dose for Port Hope Conversion Facility it details if it's "Site 1" or "Site 2":

df_npf[(df_npf['Facility Name'] == 'Port Hope Conversion Facility') & (df_npf['Substance Name (English)'] == 'Estimated public dose (see footnote)')].sort_values(by = ['Footnotes', 'Year'])


df_npf.drop(columns=['NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude', 'Substance Name (French) | Nom de substance (Français)', 'Footnotes'], inplace=True)
df_npf.head()


# Aggregating Port Hope Site 1 & 2 Estimated Public Dose to be able to have a same value accross the years to plot:

df_npf = df_npf.groupby(['Year', 'Facility Name', 'Substance Name (English)', 'Units'],as_index=False).agg({'Stack Emissions': 'sum', 'Direct Discharge': 'sum'})
df_npf[(df_npf['Facility Name'] == 'Port Hope Conversion Facility') & (df_npf['Substance Name (English)'] == 'Estimated public dose (see footnote)')]
df_npf.head()


# I'm saving the clean dataframe to do a dashboard in Tableau.

df_npf.to_csv(".\Datasets\df_npf.csv", index=True, header=True)


df_npf['Substance Name (English)'].unique()

array(['Estimated public dose (see footnote)', 'Uranium', 'Radium-226',
       'Cobalt-60', 'Iodine-125', 'Iodine-131', 'Xenon-133', 'Xenon-135',
       'Xenon-135m', 'Elemental Tritium (HT)', 'Tritium (HTO)'],
      dtype=object)


df_npf_epd = df_npf[df_npf['Substance Name (English)'] == 'Estimated public dose (see footnote)']
df_npf_epd.head()


# Estimated public dose is calculated incorporating all major release pathways (emissions and discharges).

plt.figure(figsize=(16,6))

year = df_npf_epd['Year'].unique()

for facility in df_npf_epd['Facility Name'].unique():
    plt.plot(year, df_npf_epd[df_npf_epd['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Estimated public dose [mSv/a]', size=12)
plt.legend(df_npf_epd['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_npf_uranium = df_npf[df_npf['Substance Name (English)'] == 'Uranium']
df_npf_uranium.head()


plt.figure(figsize=(16,6))

year = df_npf_uranium['Year'].unique()

for facility in df_npf_uranium['Facility Name'].unique():
    plt.plot(year, df_npf_uranium[df_npf_uranium['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Uranium - Stack Emissions [kg]', size=12)
plt.legend(df_npf_uranium['Facility Name'].unique(), loc='upper center')
plt.grid()

plt.show()


# Plotting without Port Hope to take a look at the rest:

df_npf_uranium2 = df_npf_uranium[df_npf_uranium['Facility Name'] != 'Port Hope Conversion Facility']

plt.figure(figsize=(16,6))

year = df_npf_uranium2['Year'].unique()

for facility in df_npf_uranium2['Facility Name'].unique():
    plt.plot(year, df_npf_uranium2[df_npf_uranium2['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Uranium - Stack Emissions [kg]', size=12)
plt.legend(df_npf_uranium2['Facility Name'].unique(), loc='upper center')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_npf_uranium['Year'].unique()

for facility in df_npf_uranium['Facility Name'].unique():
    plt.plot(year, df_npf_uranium[df_npf_uranium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Uranium - Direct Discharge [kg]', size=12)
plt.legend(df_npf_uranium['Facility Name'].unique(), loc='upper center')
plt.grid()

plt.show()


df_npf_radium = df_npf[df_npf['Substance Name (English)'] == 'Radium-226']
df_npf_radium.head()


# No Stack Emissions reported (addressed in question (2)).

plt.figure(figsize=(16,6))

plt.plot(df_npf_radium['Year'], df_npf_radium['Direct Discharge'])
plt.xticks(df_npf_radium['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Radium-226 - Direct Discharge [MBq]', size=12)
plt.legend(df_npf_radium['Facility Name'], loc='upper left')
plt.grid()

plt.show()


df_npf_HT = df_npf[df_npf['Substance Name (English)'] == 'Elemental Tritium (HT)']
df_npf_HT.head()


# No Direct Discharge reported (addressed in question (3)).

plt.figure(figsize=(16,6))

plt.plot(df_npf_HT['Year'], df_npf_HT['Stack Emissions'])
plt.xticks(df_npf_HT['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Elemental Tritium (HT) - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_HT['Facility Name'], loc='upper right')
plt.grid()

plt.show()


df_npf_HTO = df_npf[df_npf['Substance Name (English)'] == 'Tritium (HTO)']
df_npf_HTO.head()


# No Direct Discharge reported (addressed in question (3)).

plt.figure(figsize=(16,6))

plt.plot(df_npf_HTO['Year'], df_npf_HTO['Stack Emissions'])
plt.xticks(df_npf_HTO['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Tritium (HTO) - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_HTO['Facility Name'], loc='upper right')
plt.grid()

plt.show()


df_npf_cobalt = df_npf[df_npf['Substance Name (English)'] == 'Cobalt-60']
df_npf_cobalt.head()


# No Direct Discharge reported (addressed in question (3)). Zero value for 2020 addressed in question (4).

plt.figure(figsize=(16,6))

plt.plot(df_npf_cobalt['Year'], df_npf_cobalt['Stack Emissions'])
plt.xticks(df_npf_cobalt['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Cobalt-60 - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_cobalt['Facility Name'], loc='upper right')
plt.grid()

plt.show()


df_npf_i125 = df_npf[df_npf['Substance Name (English)'] == 'Iodine-125']
df_npf_i125.head()


# No Direct Discharge reported (addressed in question (3)). Zero value for 2018 to 2021 addressed in question (4).

plt.figure(figsize=(16,6))

plt.plot(df_npf_i125['Year'], df_npf_i125['Stack Emissions'])
plt.xticks(df_npf_i125['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Iodine-125 - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_i125['Facility Name'], loc='upper right')
plt.grid()

plt.show()


df_npf_i131 = df_npf[df_npf['Substance Name (English)'] == 'Iodine-131']
df_npf_i131.head()


# No Direct Discharge reported (addressed in question (3)). Zero value for 2019 to 2021 addressed in question (4).

plt.figure(figsize=(16,6))

plt.plot(df_npf_i131['Year'], df_npf_i131['Stack Emissions'])
plt.xticks(df_npf_i131['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Iodine-131 - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_i131['Facility Name'], loc='upper right')
plt.grid()

plt.show()


df_npf_x133 = df_npf[df_npf['Substance Name (English)'] == 'Xenon-133']
df_npf_x133.head()


# No Direct Discharge reported (addressed in question (3)). Zero value for 2017 to 2021 addressed in question (4).

plt.figure(figsize=(16,6))

plt.plot(df_npf_x133['Year'], df_npf_x133['Stack Emissions'])
plt.xticks(df_npf_x133['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Xenon-133 - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_x133['Facility Name'], loc='upper right')
plt.grid()

plt.show()


df_npf_x135 = df_npf[df_npf['Substance Name (English)'] == 'Xenon-135']
df_npf_x135.head()


# No Direct Discharge reported (addressed in question (3)). Zero value for 2017 to 2021 addressed in question (4).

plt.figure(figsize=(16,6))

plt.plot(df_npf_x135['Year'], df_npf_x135['Stack Emissions'])
plt.xticks(df_npf_x135['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Xenon-135 - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_x135['Facility Name'], loc='upper right')
plt.grid()

plt.show()


df_npf_x135m = df_npf[df_npf['Substance Name (English)'] == 'Xenon-135m']
df_npf_x135m.head()


# No Direct Discharge reported (addressed in question (3)). Zero value for 2017 to 2021 addressed in question (4).

plt.figure(figsize=(16,6))

plt.plot(df_npf_x135m['Year'], df_npf_x135m['Stack Emissions'])
plt.xticks(df_npf_x135m['Year'])
plt.xlabel('Year', size=16)
plt.ylabel('Xenon-135m - Stack Emissions [GBq]', size=12)
plt.legend(df_npf_x135m['Facility Name'], loc='upper right')
plt.grid()

plt.show()


facilities = df_npf['Facility Name'].unique()

for f in facilities:
    df = df_npf[df_npf['Facility Name'] == f]
    print(f,'\n')
    subs = df['Substance Name (English)'].unique()
    for s in subs:
        df2 = df[df['Substance Name (English)'] == s]
        fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(10,8))
        fig.subplots_adjust(hspace=0.5)
    
        ax1.plot(df2['Year'], df2['Stack Emissions'], color='green')
        ax1.set_title(s + ' - Stack Emissions', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax1.set_xlabel('Year', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax1.grid()
    
        ax2.plot(df2['Year'], df2['Direct Discharge'], color='red')
        ax2.set_title(s + ' - Direct Discharge', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax2.set_xlabel('Year', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax2.grid()
    
        plt.show()

BWXT - Peterborough

BWXT - Toronto

Blind River Refinery

Cameco Fuel Manufacturing

Nordion - Ottawa

Port Hope Conversion Facility

SRBT


df_npf_2020 = pd.read_csv("./Datasets/2020/Nuclear Processing Facilities.csv", encoding='latin1')
df_npf_2020.head()


# I will remove 2021 from the new dataframe and compare the remaining with 2020.

df_npf_2021 = df_npf_0[df_npf_0['Year | Année'] != 2021].reset_index(drop = True)
df_npf_2021.head()


# I will concatenate both dataframes & keep the not duplicates to see the differences. This produces a df with 2021's values first, and 2020's values after:

df = pd.concat([df_npf_2021,df_npf_2020]).drop_duplicates(keep=False)
df

# 2 changes only.


df_cnl = pd.read_csv("./Datasets/Canadian Nuclear Laboratories.csv")
df_cnl


# I'm creating a copy because I will need it later.

df_cnl_0 = df_cnl.copy()


df_cnl["Facility Name | Nom de l'installation"].unique()

array(['Chalk River Laboratories', 'Whiteshell Laboratories',
       'Port Granby Project', 'Port Hope Project', 'Douglas Point',
       'Nuclear Power Demonstration'], dtype=object)


# Renaming columns to English only:

df_cnl.rename(columns={'Year | Année': 'Year', 'NPRI ID | ID INRP': 'NPRI ID','Company Name | Raison Sociale':'Company Name',"Facility Name | Nom de l'installation":'Facility Name', 'City | Ville':'City', 'CSD | SDR':'CSD','CA or CMA | AR ou RMR':'CA or CMA', 'Economic Region | Région économique':'Economic Region','Province | Province':'Province', 'Latitude | Latitude':'Latitude', 'Longitude | Longitude':'Longitude','Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)','Units | Unités':'Units', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge','Footnotes | Notes de bas de page':'Footnotes'}, inplace = True)
df_cnl.head()


# Stack Emission column first:

df_cnl_miss_stack = df_cnl[df_cnl['Stack Emissions'] == 'NRM | NRS']
df_cnl_miss_stack[['Year','Facility Name', 'Substance Name (English)', 'Stack Emissions']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


# Direct Discharge column next:

df_cnl_miss_discharge = df_cnl[df_cnl['Direct Discharge'] == 'NRM | NRS']
df_cnl_miss_discharge[['Year','Facility Name', 'Substance Name (English)', 'Direct Discharge']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


# I noticed some values are "0.00E+00".

df_cnl[df_cnl['Stack Emissions'] == '0.00E+00'][['Year','Facility Name', 'Substance Name (English)', 'Stack Emissions', 'Direct Discharge']].sort_values(by =['Facility Name', 'Substance Name (English)', 'Year'])


df_cnl_geography = df_cnl[['Facility Name', 'NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude']]
df_cnl_geography


# Cleaning the geography dataframe:

df_cnl_geography.drop_duplicates(inplace=True)
df_cnl_geography = df_cnl_geography.reset_index(drop=True)
df_cnl_geography


# Cleaning the NRM (Not required to monitor) values so I can convert the columns into numeric for later plotting:

df_cnl['Stack Emissions'].replace('NRM | NRS', 0, inplace=True)
df_cnl['Direct Discharge'].replace('NRM | NRS', 0, inplace=True)


# Replacing the "<0.01" to "0.01" to be able to transform into numeric, and correcting the "3.08+10" to "3.08E+10" & "4.43+07" to "4.43E+07" (rows 29 & 30).

df_cnl['Stack Emissions'].replace('<0.01', 0.01, inplace=True)
df_cnl['Direct Discharge'].replace('3.08+10', 3.08E+10, inplace=True)
df_cnl['Direct Discharge'].replace('4.43+07', 4.43E+07, inplace=True)


# Converted columns to numeric for plotting:

df_cnl['Stack Emissions'] = pd.to_numeric(df_cnl['Stack Emissions'])
df_cnl['Direct Discharge'] = pd.to_numeric(df_cnl['Direct Discharge'])


# I noticed "Port Hope Project" has 4 extra values in 2017 & 2018 for Radium-226 & Uranium:

df_cnl[(df_cnl['Year'].isin([2017, 2018])) & (df_cnl['Substance Name (English)'].isin(['Radium-226', 'Uranium'])) & (df_cnl['Facility Name'] == 'Port Hope Project')].sort_values(['Substance Name (English)'])[['Year', 'Facility Name', 'Substance Name (English)', 'Units', 'Stack Emissions', 'Direct Discharge', 'Footnotes']]

# Note that the 0.0 values were "NRM" before I changed them.
# Footnote explains: Releases from non-routine operations.


df_cnl.drop(columns=['NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude', 'Substance Name (French) | Nom de substance (Français)', 'Footnotes'], inplace=True)
df_cnl.head()


# I'm going to aggregate the extra Port Hope values, to be able to plot:

df_cnl = df_cnl.groupby(['Year', 'Facility Name', 'Substance Name (English)', 'Units'],as_index=False).agg({'Stack Emissions': 'sum', 'Direct Discharge': 'sum'})
df_cnl[(df_cnl['Year'].isin([2017, 2018])) & (df_cnl['Substance Name (English)'].isin(['Radium-226', 'Uranium'])) & (df_cnl['Facility Name'] == 'Port Hope Project')]
df_cnl.head()


# I'm saving the clean dataframe to do a dashboard in Tableau.

df_cnl.to_csv(".\Datasets\df_cnl.csv", index=True, header=True)


df_cnl['Substance Name (English)'].unique()

array(['Argon-41', 'Carbon-14', 'Elemental Tritium (HT)',
       'Estimated public dose (see footnote)', 'Iodine-125', 'Iodine-131',
       'Particulate gross alpha', 'Particulate gross beta',
       'Strontium-90', 'Total noble gases', 'Tritium (HTO)', 'Xenon-133',
       'Radium-226', 'Uranium', 'Cesium-137', 'Americium-241',
       'Plutonium-238', 'Plutonium-239/240', 'Uranium-total'],
      dtype=object)


df_cnl_c14 = df_cnl[df_cnl['Substance Name (English)'] == 'Carbon-14']
df_cnl_c14


# No Direct Discharge reported (addressed in question (3)). Removing Douglas to be able to plot.

df_cnl_c14 = df_cnl_c14[df_cnl_c14['Facility Name'] != 'Douglas Point']

plt.figure(figsize=(16,6))

year = df_cnl_c14['Year'].unique()

for facility in df_cnl_c14['Facility Name'].unique():
    plt.plot(year, df_cnl_c14[df_cnl_c14['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Carbon-14 - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_c14['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_cnl_epd = df_cnl[df_cnl['Substance Name (English)'] == 'Estimated public dose (see footnote)']
df_cnl_epd.head()


# Whiteshell Laboratories, Port Granby Project, Port Hope Project, Nuclear Power Demonstration are missing 2013. I will have to plot them separately.

# No Direct Discharge reported (addressed in question (3)).

df_cnl_epd2 = df_cnl_epd[df_cnl_epd['Facility Name'].isin(['Chalk River Laboratories', 'Douglas Point'])]

plt.figure(figsize=(16,6))

year = df_cnl_epd2['Year'].unique()

for facility in df_cnl_epd2['Facility Name'].unique():
    plt.plot(year, df_cnl_epd2[df_cnl_epd2['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Estimated public dose - Stack Emissions [mSv/a]', size=12)
plt.legend(df_cnl_epd2['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


# No Direct Discharge reported (addressed in question (3)).

df_cnl_epd3 = df_cnl_epd[df_cnl_epd['Facility Name'].isin(['Whiteshell Laboratories', 'Port Granby Project', 'Port Hope Project', 'Nuclear Power Demonstration'])]

plt.figure(figsize=(16,6))

year = df_cnl_epd3['Year'].unique()

for facility in df_cnl_epd3['Facility Name'].unique():
    plt.plot(year, df_cnl_epd3[df_cnl_epd3['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Estimated public dose - Stack Emissions [mSv/a]', size=12)
plt.legend(df_cnl_epd3['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_cnl_alpha = df_cnl[df_cnl['Substance Name (English)'] == 'Particulate gross alpha']
df_cnl_alpha.head()


# Douglas Point starts in 2016. I will have to plot it separately.

# Chalk River Laboratories Stack Emissions are 0, but it was previously 'NRM' (addressed in question (2)).

df_cnl_alpha2 = df_cnl_alpha[df_cnl_alpha['Facility Name'] != 'Douglas Point']
                             
plt.figure(figsize=(16,6))

year = df_cnl_alpha2['Year'].unique()

for facility in df_cnl_alpha2['Facility Name'].unique():
    plt.plot(year, df_cnl_alpha2[df_cnl_alpha2['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross alpha - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_alpha2['Facility Name'].unique(), loc='center left')
plt.grid()

plt.show()


df_cnl_alpha3 = df_cnl_alpha[df_cnl_alpha['Facility Name'] == 'Douglas Point']
                             
plt.figure(figsize=(16,6))

year = df_cnl_alpha3['Year'].unique()

for facility in df_cnl_alpha3['Facility Name'].unique():
    plt.plot(year, df_cnl_alpha3[df_cnl_alpha3['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross alpha - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_alpha3['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_cnl_alpha2['Year'].unique()

for facility in df_cnl_alpha2['Facility Name'].unique():
    plt.plot(year, df_cnl_alpha2[df_cnl_alpha2['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross alpha - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_alpha2['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_cnl_alpha3['Year'].unique()

for facility in df_cnl_alpha3['Facility Name'].unique():
    plt.plot(year, df_cnl_alpha3[df_cnl_alpha3['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross alpha - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_alpha3['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_beta = df_cnl[df_cnl['Substance Name (English)'] == 'Particulate gross beta']
df_cnl_beta.head()


# Chalk River Laboratories & Douglas Point Stack Emissions are 0, but it was previously 'NRM' (addressed in question (2)).

plt.figure(figsize=(16,6))

year = df_cnl_beta['Year'].unique()

for facility in df_cnl_beta['Facility Name'].unique():
    plt.plot(year, df_cnl_beta[df_cnl_beta['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross beta - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_beta['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


plt.figure(figsize=(16,6))

year = df_cnl_beta['Year'].unique()

for facility in df_cnl_beta['Facility Name'].unique():
    plt.plot(year, df_cnl_beta[df_cnl_beta['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross beta - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_beta['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


# I'm plotting without Chalk River to see the rest.

df_cnl_beta2 = df_cnl_beta[df_cnl_beta['Facility Name'] != 'Chalk River Laboratories']

plt.figure(figsize=(16,6))

year = df_cnl_beta['Year'].unique()

for facility in df_cnl_beta2['Facility Name'].unique():
    plt.plot(year, df_cnl_beta2[df_cnl_beta2['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Particulate gross beta - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_beta2['Facility Name'].unique(), loc='best')
plt.grid()

plt.show()


df_cnl_radium = df_cnl[df_cnl['Substance Name (English)'] == 'Radium-226']
df_cnl_radium.head()


# No Stack Emissions reported (addressed in question (2)).

plt.figure(figsize=(16,6))

year = df_cnl_radium['Year'].unique()

for facility in df_cnl_radium['Facility Name'].unique():
    plt.plot(year, df_cnl_radium[df_cnl_radium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Radium-226 - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_radium['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_strontium = df_cnl[df_cnl['Substance Name (English)'] == 'Strontium-90']
df_cnl_strontium.head()


# No Stack Emissions reported (addressed in question (2)).

plt.figure(figsize=(16,6))

year = df_cnl_strontium['Year'].unique()

for facility in df_cnl_strontium['Facility Name'].unique():
    plt.plot(year, df_cnl_strontium[df_cnl_strontium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Strontium-90 - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_strontium['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_cnl_hto = df_cnl[df_cnl['Substance Name (English)'] == 'Tritium (HTO)']
df_cnl_hto.head()


plt.figure(figsize=(16,6))

year = df_cnl_hto['Year'].unique()

for facility in df_cnl_hto['Facility Name'].unique():
    plt.plot(year, df_cnl_hto[df_cnl_hto['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Tritium (HTO) - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_hto['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


# I'm plotting without Chalk River to see the rest.

df_cnl_hto2 = df_cnl_hto[df_cnl_hto['Facility Name'] != 'Chalk River Laboratories']

plt.figure(figsize=(16,6))

year = df_cnl_hto2['Year'].unique()

for facility in df_cnl_hto2['Facility Name'].unique():
    plt.plot(year, df_cnl_hto2[df_cnl_hto2['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Tritium (HTO) - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_hto2['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


# No Direct Discharge reported for Whiteshell Laboratories (NRM, addressed in question (3)).

plt.figure(figsize=(16,6))

year = df_cnl_hto['Year'].unique()

for facility in df_cnl_hto['Facility Name'].unique():
    plt.plot(year, df_cnl_hto[df_cnl_hto['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Tritium (HTO) - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_hto['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


# I'm plotting without Chalk River to see the rest.

plt.figure(figsize=(16,6))

year = df_cnl_hto2['Year'].unique()

for facility in df_cnl_hto2['Facility Name'].unique():
    plt.plot(year, df_cnl_hto2[df_cnl_hto2['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Tritium (HTO) - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_hto2['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_cnl_uranium = df_cnl[df_cnl['Substance Name (English)'] == 'Uranium']
df_cnl_uranium.head()


# No Stack Emissions reported (addressed in question (2)).

plt.figure(figsize=(16,6))

year = df_cnl_uranium['Year'].unique()

for facility in df_cnl_uranium['Facility Name'].unique():
    plt.plot(year, df_cnl_uranium[df_cnl_uranium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Uranium - Direct Discharge [kg]', size=12)
plt.legend(df_cnl_uranium['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_a241 = df_cnl[df_cnl['Substance Name (English)'] == 'Americium-241']
df_cnl_a241.head()


# No Stack Emissions reported (addressed in question (2)). Missing Direct Discharge values from 2013 to 2016.

plt.figure(figsize=(16,6))

year = df_cnl_a241['Year'].unique()

for facility in df_cnl_a241['Facility Name'].unique():
    plt.plot(year, df_cnl_a241[df_cnl_a241['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Americium-241 - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_a241['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_argon = df_cnl[df_cnl['Substance Name (English)'] == 'Argon-41']
df_cnl_argon.head()


# No Direct Discharge reported (addressed in question (3)). Zero values for 2019 - 2021 (not NRM, addressed in question (4))

plt.figure(figsize=(16,6))

year = df_cnl_argon['Year'].unique()

for facility in df_cnl_argon['Facility Name'].unique():
    plt.plot(year, df_cnl_argon[df_cnl_argon['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Argon-41 - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_argon['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_cesium = df_cnl[df_cnl['Substance Name (English)'] == 'Cesium-137']
df_cnl_cesium.head()


# No Stack Emissions reported (addressed in question (2)).

plt.figure(figsize=(16,6))

year = df_cnl_cesium['Year'].unique()

for facility in df_cnl_cesium['Facility Name'].unique():
    plt.plot(year, df_cnl_cesium[df_cnl_cesium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Cesium-137 - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_cesium['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_cnl_ht = df_cnl[df_cnl['Substance Name (English)'] == 'Elemental Tritium (HT)']
df_cnl_ht.head()


# No Direct Discharge reported (addressed in question (3)).

plt.figure(figsize=(16,6))

year = df_cnl_ht['Year'].unique()

for facility in df_cnl_ht['Facility Name'].unique():
    plt.plot(year, df_cnl_ht[df_cnl_ht['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Elemental Tritium (HT) - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_ht['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_i125 = df_cnl[df_cnl['Substance Name (English)'] == 'Iodine-125']
df_cnl_i125.head()


# No Direct Discharge reported (addressed in question (3)).

plt.figure(figsize=(16,6))

year = df_cnl_i125['Year'].unique()

for facility in df_cnl_i125['Facility Name'].unique():
    plt.plot(year, df_cnl_i125[df_cnl_i125['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Iodine-125 - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_i125['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_i131 = df_cnl[df_cnl['Substance Name (English)'] == 'Iodine-131']
df_cnl_i131.head()


# No Direct Discharge reported (addressed in question (3)).

plt.figure(figsize=(16,6))

year = df_cnl_i131['Year'].unique()

for facility in df_cnl_i131['Facility Name'].unique():
    plt.plot(year, df_cnl_i131[df_cnl_i131['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Iodine-131 - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_i131['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_cnl_p238 = df_cnl[df_cnl['Substance Name (English)'] == 'Plutonium-238']
df_cnl_p238.head()


# No Stack Emissions reported (addressed in question (2)).

plt.figure(figsize=(16,6))

year = df_cnl_p238['Year'].unique()

for facility in df_cnl_p238['Facility Name'].unique():
    plt.plot(year, df_cnl_p238[df_cnl_p238['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Plutonium-238 - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_p238['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_p239 = df_cnl[df_cnl['Substance Name (English)'] == 'Plutonium-239/240']
df_cnl_p239.head()


# No Stack Emissions reported (addressed in question (2)).

plt.figure(figsize=(16,6))

year = df_cnl_p239['Year'].unique()

for facility in df_cnl_p239['Facility Name'].unique():
    plt.plot(year, df_cnl_p239[df_cnl_p239['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Plutonium-239/240 - Direct Discharge [Bq]', size=12)
plt.legend(df_cnl_p239['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_cnl_noble = df_cnl[df_cnl['Substance Name (English)'] == 'Total noble gases']
df_cnl_noble.head()


# No Direct Discharge reported (addressed in question (3)).

plt.figure(figsize=(16,6))

year = df_cnl_noble['Year'].unique()

for facility in df_cnl_noble['Facility Name'].unique():
    plt.plot(year, df_cnl_noble[df_cnl_noble['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Total noble gases - Stack Emissions [Bq-MeV]', size=12)
plt.legend(df_cnl_noble['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_cnl_xenon = df_cnl[df_cnl['Substance Name (English)'] == 'Xenon-133']
df_cnl_xenon.head()


# No Direct Discharge reported (addressed in question (3)).

plt.figure(figsize=(16,6))

year = df_cnl_xenon['Year'].unique()

for facility in df_cnl_xenon['Facility Name'].unique():
    plt.plot(year, df_cnl_xenon[df_cnl_xenon['Facility Name'] == facility]['Stack Emissions'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Xenon-133 - Stack Emissions [Bq]', size=12)
plt.legend(df_cnl_xenon['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


facilities = df_cnl['Facility Name'].unique()

for f in facilities:
    df = df_cnl[df_cnl['Facility Name'] == f]
    print(f,'\n')
    subs = df['Substance Name (English)'].unique()
    for s in subs:
        df2 = df[df['Substance Name (English)'] == s]
        fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(10,8))
        fig.subplots_adjust(hspace=0.5)
    
        ax1.plot(df2['Year'], df2['Stack Emissions'], color='green')
        ax1.set_title(s + ' - Stack Emissions', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax1.set_xlabel('Year', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax1.grid()
    
        ax2.plot(df2['Year'], df2['Direct Discharge'], color='red')
        ax2.set_title(s + ' - Direct Discharge', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax2.set_xlabel('Year', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        ax2.grid()
    
        plt.show()

Chalk River Laboratories

Douglas Point

Nuclear Power Demonstration

Port Granby Project

Port Hope Project

Whiteshell Laboratories


df_cnl_2020 = pd.read_csv("./Datasets/2020/Canadian Nuclear Laboratories.csv", encoding='latin1')
df_cnl_2020.head()


# I will leave only the essential columns, as I saw errors in the other ones (in NPRI ID for example) and those are not important.

df_cnl_2021 = df_cnl_0[['Year | Année', "Facility Name | Nom de l'installation", 'Substance Name (English) | Nom de substance (Anglais)', 'Stack Emissions | Émissions de cheminées',
       'Direct Discharge | Évacuations directes']]
df_cnl_2021.rename(columns={'Year | Année': 'Year', "Facility Name | Nom de l'installation":'Facility Name', 'Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge'}, inplace = True)
df_cnl_2021.head()


# I will remove 2021 from the new dataframe and compare the remaining with 2020.

df_cnl_2021 = df_cnl_2021[df_cnl_2021['Year'] != 2021].reset_index(drop = True)
df_cnl_2021.head()


# I will leave only the essential columns, as I saw errors in the other ones (in NPRI ID for example) and those are not important.

df_cnl_2020 = df_cnl_2020[['Year | Année', "Facility Name | Nom de l'installation", 'Substance Name (English) | Nom de substance (Anglais)', 'Stack Emissions | Émissions de cheminées',
       'Direct Discharge | Évacuations directes']]
df_cnl_2020.rename(columns={'Year | Année': 'Year', "Facility Name | Nom de l'installation":'Facility Name', 'Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge'}, inplace = True)
df_cnl_2020.head()


# I will concatenate both dataframes & keep the not duplicates to see the differences. This produces a df with 2021's values first, and 2020's values after:

df = pd.concat([df_cnl_2021,df_cnl_2020]).drop_duplicates(keep=False)
df = df.sort_values(by=['Facility Name', 'Substance Name (English)', 'Year'])
df

# I will explore the differences:


df_cnl_2021[(df_cnl_2021['Facility Name'] == 'Chalk River Laboratories') & (df_cnl_2021['Substance Name (English)'] == 'Iodine-125')]


df_cnl_2020[(df_cnl_2020['Facility Name'] == 'Chalk River Laboratories') & (df_cnl_2020['Substance Name (English)'] == 'Iodine-125')]


df_cnl_2021[(df_cnl_2021['Facility Name'] == 'Chalk River Laboratories') & (df_cnl_2021['Substance Name (English)'] == 'Strontium-90')]


df_cnl_2020[(df_cnl_2020['Facility Name'] == 'Chalk River Laboratories') & (df_cnl_2020['Substance Name (English)'] == 'Strontium-90')]


df[(df['Substance Name (English)'] != 'Iodine-125') & (df['Substance Name (English)'] != 'Strontium-90')]


df_umm = pd.read_csv("./Datasets/Uranium Mines and Mills.csv")
df_umm


# I'm creating a copy because I will need it later.

df_umm_0 = df_umm.copy()


df_umm["Facility Name | Nom de l'installation"].unique()

array(['Rabbit Lake', 'Key Lake', 'McArthur River', 'Cigar Lake',
       'McClean Lake'], dtype=object)


# All Stack Emissions are NRM (Not Required to Monitor). So I will focus this analysis on Direct Discharge only.

df_umm['Stack Emissions | Émissions de cheminées'].unique()

array(['NRM | NRS'], dtype=object)


# Renaming columns to English only:

df_umm.rename(columns={'Year | Année': 'Year', 'NPRI ID | ID INRP': 'NPRI ID','Company Name | Raison Sociale':'Company Name',"Facility Name | Nom de l'installation":'Facility Name', 'City | Ville':'City', 'CSD | SDR':'CSD','CA or CMA | AR ou RMR':'CA or CMA', 'Economic Region | Région économique':'Economic Region','Province | Province':'Province', 'Latitude | Latitude':'Latitude', 'Longitude | Longitude':'Longitude','Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)','Units | Unités':'Units', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge','Footnotes | Notes de bas de page':'Footnotes'}, inplace = True)
df_umm.head()


# After checking Direct Discharge, I noticed there is a value "DL" & a 0 value:

df_umm[(df_umm['Direct Discharge'] == 'DL') | (df_umm['Direct Discharge'] == '0.0')]


# Cleaning the geography dataframe:

df_umm_geography = df_umm[['Facility Name', 'NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude']]
df_umm_geography.drop_duplicates(inplace=True)
df_umm_geography = df_umm_geography.reset_index(drop=True)
df_umm_geography


# Cleaning the DL value so I can convert the column into numeric for later plotting:

df_umm['Direct Discharge'].replace('DL', 0, inplace=True)


# Converted columns to numeric for plotting:

df_umm['Direct Discharge'] = pd.to_numeric(df_umm['Direct Discharge'])


df_umm.drop(columns=['NPRI ID', 'Company Name', 'City', 'CSD', 'CA or CMA', 'Economic Region', 'Province', 'Latitude', 'Longitude', 'Substance Name (French) | Nom de substance (Français)', 'Stack Emissions', 'Footnotes', 'Unnamed: 17', 'Unnamed: 18'], inplace=True)
df_umm.head()


# First, I'm saving the clean dataframe to do a dashboard in Tableau.

df_umm.to_csv(".\Datasets\df_umm.csv", index=True, header=True)


df_umm_uranium = df_umm[df_umm['Substance Name (English)'] == 'Uranium']
df_umm_uranium.head()


plt.figure(figsize=(16,6))

year = df_umm_uranium['Year'].unique()

for facility in df_umm_uranium['Facility Name'].unique():
    plt.plot(year, df_umm_uranium[df_umm_uranium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Uranium - Direct Discharge [kg]', size=12)
plt.legend(df_umm_uranium['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


# plotting without Rabbit Lake to have a better look at the rest.

df_umm_uranium_2 = df_umm_uranium[df_umm_uranium['Facility Name'] != 'Rabbit Lake']

plt.figure(figsize=(16,6))

year = df_umm_uranium_2['Year'].unique()

for facility in df_umm_uranium_2['Facility Name'].unique():
    plt.plot(year, df_umm_uranium_2[df_umm_uranium_2['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Uranium - Direct Discharge [kg]', size=12)
plt.legend(df_umm_uranium_2['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_umm_thorium = df_umm[df_umm['Substance Name (English)'] == 'Thorium-230']
df_umm_thorium.head()


plt.figure(figsize=(16,6))

year = df_umm_thorium['Year'].unique()

for facility in df_umm_thorium['Facility Name'].unique():
    plt.plot(year, df_umm_thorium[df_umm_thorium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Thorium-230 - Direct Discharge [MBq]', size=12)
plt.legend(df_umm_thorium['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_umm_radium = df_umm[df_umm['Substance Name (English)'] == 'Radium-226']
df_umm_radium.head()


plt.figure(figsize=(16,6))

year = df_umm_radium['Year'].unique()

for facility in df_umm_radium['Facility Name'].unique():
    plt.plot(year, df_umm_radium[df_umm_radium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Radium-226 - Direct Discharge [MBq]', size=12)
plt.legend(df_umm_radium['Facility Name'].unique(), loc='upper right')
plt.grid()

plt.show()


df_umm_lead = df_umm[df_umm['Substance Name (English)'] == 'Lead-210']
df_umm_lead.head()


plt.figure(figsize=(16,6))

year = df_umm_lead['Year'].unique()

for facility in df_umm_lead['Facility Name'].unique():
    plt.plot(year, df_umm_lead[df_umm_lead['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Lead-210 - Direct Discharge [MBq]', size=12)
plt.legend(df_umm_lead['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


df_umm_polonium = df_umm[df_umm['Substance Name (English)'] == 'Polonium-210']
df_umm_polonium.head()


plt.figure(figsize=(16,6))

year = df_umm_polonium['Year'].unique()

for facility in df_umm_polonium['Facility Name'].unique():
    plt.plot(year, df_umm_polonium[df_umm_polonium['Facility Name'] == facility]['Direct Discharge'])
plt.xticks(year)
plt.xlabel('Year', size=16)
plt.ylabel('Polonium-210 - Direct Discharge [MBq]', size=12)
plt.legend(df_umm_polonium['Facility Name'].unique(), loc='upper left')
plt.grid()

plt.show()


facilities = df_umm['Facility Name'].unique()

for f in facilities:
    df = df_umm[df_umm['Facility Name'] == f]
    print(f,'\n')
    subs = df['Substance Name (English)'].unique()
    for s in subs:
        df2 = df[df['Substance Name (English)'] == s]
        plt.figure(figsize=(12,4))
        plt.plot(df2['Year'], df2['Direct Discharge'], color='red')
        plt.title(s + ' - Direct Discharge', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        plt.xlabel('Year', fontdict={'fontsize': 14, 'fontweight': 'medium'})
        plt.grid()
    
        plt.show()

Rabbit Lake

Key Lake

McArthur River

Cigar Lake

McClean Lake


df_umm_2020 = pd.read_csv("./Datasets/2020/Uranium Mines and Mills.csv")
df_umm_2020.head()


# I will leave only the essential columns.

df_umm_2021 = df_umm_0[['Year | Année', "Facility Name | Nom de l'installation", 'Substance Name (English) | Nom de substance (Anglais)', 'Stack Emissions | Émissions de cheminées',
       'Direct Discharge | Évacuations directes']]
df_umm_2021.rename(columns={'Year | Année': 'Year', "Facility Name | Nom de l'installation":'Facility Name', 'Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)', 'Stack Emissions | Émissions de cheminées':'Stack Emissions','Direct Discharge | Évacuations directes':'Direct Discharge'}, inplace = True)
df_umm_2021.head()


# I will remove 2021 from the new dataframe and compare the remaining with 2020.

df_umm_2021 = df_umm_2021[df_umm_2021['Year'] != 2021].reset_index(drop = True)
df_umm_2021.head()


# I will leave only the essential columns for 2020 too.

df_umm_2020 = df_umm_2020[['Year | Annee', "Facility Name | Nom de l'installation", 'Substance Name (English) | Nom de substance (Anglais)', 'Stack Emissions | Emissions de cheminees',
       'Direct Discharge | Evacuations directes']]
df_umm_2020.rename(columns={'Year | Annee': 'Year', "Facility Name | Nom de l'installation":'Facility Name', 'Substance Name (English) | Nom de substance (Anglais)':'Substance Name (English)', 'Stack Emissions | Emissions de cheminees':'Stack Emissions','Direct Discharge | Evacuations directes':'Direct Discharge'}, inplace = True)
df_umm_2020.head()


# I will concatenate both dataframes & keep the not duplicates to see the differences. This produces a df with 2021's values first, and 2020's values after:

df = pd.concat([df_umm_2021,df_umm_2020]).drop_duplicates(keep=False)
df = df.sort_values(by=['Facility Name', 'Substance Name (English)', 'Year'])
df

# No errors at all.

	Year \| Année	NPRI ID \| ID INRP	Company Name \| Raison Sociale	Facility Name \| Nom de l'installation	City \| Ville	CSD \| SDR	CA or CMA \| AR ou RMR	Economic Region \| Région économique	Province \| Province	Latitude \| Latitude	Longitude \| Longitude	Substance Name (English) \| Nom de substance (Anglais)	Substance Name (French) \| Nom de substance (Français)	Units \| Unités	<	Stack Emissions \| Émissions de cheminées	<.1	Direct Discharge \| Évacuations directes	Footnotes \| Notes de bas de page
0	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Tritium (HTO)	Tritium (Eau tritiée)	Bq	NaN	4.43E+13	NaN	1.56E+14	NaN
1	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Carbon-14	Carbone-14	Bq	NaN	6.17E+09	NaN	6.07E+07	NaN
2	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Total noble gases	Total des gaz nobles	Bq-MeV	NaN	NRM \| NRS	NaN	NRM \| NRS	NaN
3	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Iodine-131	Iode-131	Bq	NaN	NRM \| NRS	NaN	NRM \| NRS	NaN
4	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Particulate (gross beta/gamma)	Particules (bêta brutes/gamma brutes)	Bq	NaN	5.11E+05	NaN	7.11E+07	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
494	2011	3161	Ontario Power Generation Inc.	Pickering Nuclear - B	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676	Total noble gases	Total des gaz nobles	Bq-MeV	NaN	8.40E+13	NaN	NRM \| NRS	NaN
495	2011	3161	Ontario Power Generation Inc.	Pickering Nuclear - B	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676	Iodine-131	Iode-131	Bq	NaN	8.80E+06	NaN	NRM \| NRS	NaN
496	2011	3161	Ontario Power Generation Inc.	Pickering Nuclear - B	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676	Particulate (gross beta/gamma)	Particules (bêta brutes/gamma brutes)	Bq	NaN	3.60E+06	NaN	1.40E+10	NaN
497	2011	3161	Ontario Power Generation Inc.	Pickering Nuclear - B	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676	Particulate gross alpha	Particules alpha brutes	Bq	NaN	NRM \| NRS	NaN	4.80E+07	NaN
498	2011	3163	Ontario Power Generation Inc.	Pickering Nuclear - A & B	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676	Estimated public dose (see footnote)	Dose estimée au public (voir note de bas de page)	mSv/a	NaN	0.0009	NaN	NRM \| NRS	Estimated public dose is calculated incorporat...

	Year	NPRI ID	Company Name	Facility Name	City	CSD	CA or CMA	Economic Region	Province	Latitude	Longitude	Substance Name (English)	Substance Name (French) \| Nom de substance (Français)	Units	<	Stack Emissions	<.1	Direct Discharge	Footnotes
0	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Tritium (HTO)	Tritium (Eau tritiée)	Bq	NaN	4.43E+13	NaN	1.56E+14	NaN
1	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Carbon-14	Carbone-14	Bq	NaN	6.17E+09	NaN	6.07E+07	NaN
2	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Total noble gases	Total des gaz nobles	Bq-MeV	NaN	NRM \| NRS	NaN	NRM \| NRS	NaN
3	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Iodine-131	Iode-131	Bq	NaN	NRM \| NRS	NaN	NRM \| NRS	NaN
4	2021	1445	Hydro-Québec	Gentilly-2	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569	Particulate (gross beta/gamma)	Particules (bêta brutes/gamma brutes)	Bq	NaN	5.11E+05	NaN	7.11E+07	NaN

	Facility Name	NPRI ID	Company Name	City	CSD	CA or CMA	Economic Region	Province	Latitude	Longitude
0	Gentilly-2	1445	Hydro-Québec	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569
1	Gentilly-2	1445	Hydro-Québec	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569
2	Gentilly-2	1445	Hydro-Québec	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569
3	Gentilly-2	1445	Hydro-Québec	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569
4	Gentilly-2	1445	Hydro-Québec	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569
...	...	...	...	...	...	...	...	...	...	...
494	Pickering Nuclear - A & B	3161	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676
495	Pickering Nuclear - A & B	3161	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676
496	Pickering Nuclear - A & B	3161	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676
497	Pickering Nuclear - A & B	3161	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676
498	Pickering Nuclear - A & B	3163	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676

	Facility Name	NPRI ID	Company Name	City	CSD	CA or CMA	Economic Region	Province	Latitude	Longitude
0	Gentilly-2	1445	Hydro-Québec	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569
1	Point Lepreau Generating Station	1710	New Brunswick Power Corporation	Maces Bay	Musquash	Saint John	Saint John--St. Stephen	NB	45.0690	-66.4556
2	Bruce Power Site	7041	Bruce Power LP	Tiverton	Kincardine	NaN	Stratford--Bruce Peninsula	ON	44.3289	-81.5916
3	Darlington Nuclear	3163	Ontario Power Generation Inc.	Bowmanville	Clarington	Oshawa	Toronto	ON	43.8681	-78.7250
4	Pickering Nuclear - A & B	3161	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676
5	Pickering Nuclear - A & B	3163	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676

	Facility Name	NPRI ID	Company Name	City	CSD	CA or CMA	Economic Region	Province	Latitude	Longitude
0	Gentilly-2	1445	Hydro-Québec	Bécancour	NaN	NaN	NaN	QC	46.3958	-72.3569
1	Point Lepreau Generating Station	1710	New Brunswick Power Corporation	Maces Bay	Musquash	Saint John	Saint John--St. Stephen	NB	45.0690	-66.4556
2	Bruce Power Site	7041	Bruce Power LP	Tiverton	Kincardine	NaN	Stratford--Bruce Peninsula	ON	44.3289	-81.5916
3	Darlington Nuclear	3163	Ontario Power Generation Inc.	Bowmanville	Clarington	Oshawa	Toronto	ON	43.8681	-78.7250
4	Pickering Nuclear - A & B	3161	Ontario Power Generation Inc.	Pickering	Pickering	Toronto	Toronto	ON	43.8104	-79.0676

	Year	Facility Name	Substance Name (English)	Stack Emissions
239	2016	Bruce Power - B	Iodine-131	LD
484	2011	Darlington Nuclear	Particulate gross alpha	NRM \| NRS
437	2012	Darlington Nuclear	Particulate gross alpha	NRM \| NRS
455	2011	Gentilly-2	Iodine-131	LD
361	2013	Gentilly-2	Iodine-131	LD
314	2014	Gentilly-2	Iodine-131	LD
267	2015	Gentilly-2	Iodine-131	NRM \| NRS
220	2016	Gentilly-2	Iodine-131	NRM \| NRS
173	2017	Gentilly-2	Iodine-131	NRM \| NRS
126	2018	Gentilly-2	Iodine-131	NRM \| NRS
85	2019	Gentilly-2	Iodine-131	NRM \| NRS
44	2020	Gentilly-2	Iodine-131	NRM \| NRS
3	2021	Gentilly-2	Iodine-131	NRM \| NRS
266	2015	Gentilly-2	Total noble gases	NRM \| NRS
219	2016	Gentilly-2	Total noble gases	NRM \| NRS
172	2017	Gentilly-2	Total noble gases	NRM \| NRS
125	2018	Gentilly-2	Total noble gases	NRM \| NRS
84	2019	Gentilly-2	Total noble gases	NRM \| NRS
43	2020	Gentilly-2	Total noble gases	NRM \| NRS
2	2021	Gentilly-2	Total noble gases	NRM \| NRS
491	2011	Pickering Nuclear - A	Particulate gross alpha	NRM \| NRS
444	2012	Pickering Nuclear - A	Particulate gross alpha	NRM \| NRS
497	2011	Pickering Nuclear - B	Particulate gross alpha	NRM \| NRS
450	2012	Pickering Nuclear - B	Particulate gross alpha	NRM \| NRS
461	2011	Point Lepreau Generating Station	Iodine-131	NRM \| NRS
414	2012	Point Lepreau Generating Station	Iodine-131	NRM \| NRS
367	2013	Point Lepreau Generating Station	Iodine-131	NRM \| NRS
320	2014	Point Lepreau Generating Station	Iodine-131	NRM \| NRS
462	2011	Point Lepreau Generating Station	Particulate (gross beta/gamma)	NRM \| NRS
415	2012	Point Lepreau Generating Station	Particulate (gross beta/gamma)	NRM \| NRS
368	2013	Point Lepreau Generating Station	Particulate (gross beta/gamma)	NRM \| NRS
321	2014	Point Lepreau Generating Station	Particulate (gross beta/gamma)	NRM \| NRS
463	2011	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
416	2012	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
369	2013	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
322	2014	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
275	2015	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
228	2016	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
181	2017	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
134	2018	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
93	2019	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
52	2020	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
11	2021	Point Lepreau Generating Station	Particulate gross alpha	NRM \| NRS
460	2011	Point Lepreau Generating Station	Total noble gases	NRM \| NRS

	Year	Facility Name	Substance Name (English)	Direct Discharge
468	2011	Bruce Power - A	Iodine-131	NRM \| NRS
421	2012	Bruce Power - A	Iodine-131	NRM \| NRS
374	2013	Bruce Power - A	Iodine-131	NRM \| NRS
327	2014	Bruce Power - A	Iodine-131	NRM \| NRS
280	2015	Bruce Power - A	Iodine-131	NRM \| NRS
...	...	...	...	...
178	2017	Point Lepreau Generating Station	Total noble gases	NRM \| NRS
131	2018	Point Lepreau Generating Station	Total noble gases	NRM \| NRS
90	2019	Point Lepreau Generating Station	Total noble gases	NRM \| NRS
49	2020	Point Lepreau Generating Station	Total noble gases	NRM \| NRS
8	2021	Point Lepreau Generating Station	Total noble gases	NRM \| NRS

	Year	Facility Name	Substance Name (English)	Units	Stack Emissions
14	2011	Darlington Nuclear	Elemental Tritium (HT)	Bq	8.800000e+13
55	2012	Darlington Nuclear	Elemental Tritium (HT)	Bq	2.600000e+13
96	2013	Darlington Nuclear	Elemental Tritium (HT)	Bq	1.800000e+13
137	2014	Darlington Nuclear	Elemental Tritium (HT)	Bq	5.200000e+13
178	2015	Darlington Nuclear	Elemental Tritium (HT)	Bq	1.700000e+13

	Year \| Année	NPRI ID \| ID INRP	Company Name \| Raison Sociale	Facility Name \| Nom de l'installation	City \| Ville	CSD \| SDR	CA or CMA \| AR ou RMR	Economic Region \| Région économique	Province \| Province	Latitude \| Latitude	Longitude \| Longitude	Substance Name (English) \| Nom de substance (Anglais)	Substance Name (French) \| Nom de substance (Français)	Units \| Unités	Stack Emissions \| Émissions de cheminées	Direct Discharge \| Évacuations directes	Footnotes \| Notes de bas de page
0	2021	3657.0	Cameco	Blind River Refinery	Blind River	NaN	NaN	NaN	ON	46.1814	-83.0177	Uranium	Uranium	kg	3.2	2.2	NaN
1	2021	3657.0	Cameco	Blind River Refinery	Blind River	NaN	NaN	NaN	ON	46.1814	-83.0177	Radium-226	Radium-226	MBq	NRM \| NRS	2.2	NaN
2	2021	3657.0	Cameco	Blind River Refinery	Blind River	NaN	NaN	NaN	ON	46.1814	-83.0177	Estimated public dose (see footnote)	Dose estimée au public (voir note de bas de page)	mSv/a	0.009	NRM \| NRS	Estimated public dose is calculated incorporat...
3	2021	1145.0	Cameco	Port Hope Conversion Facility	Port Hope	NaN	NaN	NaN	ON	43.9437	-78.2954	Uranium	Uranium	kg	39	NRM \| NRS	NaN
4	2021	1145.0	Cameco	Port Hope Conversion Facility	Port Hope	NaN	NaN	NaN	ON	43.9437	-78.2954	Estimated public dose (see footnote)	Dose estimée au public (voir note de bas de page)	mSv/a	0.072	NRM \| NRS	Site 1, Estimated public dose is calculated in...
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
189	2013	2247.0	Nordion (Canada) Inc.	Nordion - Ottawa	Ottawa	NaN	NaN	NaN	ON	45.3408	-75.9179	Iodine-131	Iode-131	GBq	3.90E-01	NRM \| NRS	NaN
190	2013	2247.0	Nordion (Canada) Inc.	Nordion - Ottawa	Ottawa	NaN	NaN	NaN	ON	45.3408	-75.9179	Xenon-133	Xénon-133	GBq	3.07E+04	NRM \| NRS	NaN
191	2013	2247.0	Nordion (Canada) Inc.	Nordion - Ottawa	Ottawa	NaN	NaN	NaN	ON	45.3408	-75.9179	Xenon-135	Xénon-135	GBq	2.82E+04	NRM \| NRS	NaN
192	2013	2247.0	Nordion (Canada) Inc.	Nordion - Ottawa	Ottawa	NaN	NaN	NaN	ON	45.3408	-75.9179	Xenon-135m	Xénon-135m	GBq	4.34E+04	NRM \| NRS	NaN
193	2013	2247.0	Nordion (Canada) Inc.	Nordion - Ottawa	Ottawa	NaN	NaN	NaN	ON	45.3408	-75.9179	Estimated public dose (see footnote)	Dose estimée au public (voir note de bas de page)	mSv/a	0.0195	NRM \| NRS	Estimated public dose is calculated incorporat...

	Year	Facility Name	Substance Name (English)	Direct Discharge
183	2013	BWXT - Peterborough	Estimated public dose (see footnote)	NRM \| NRS
162	2014	BWXT - Peterborough	Estimated public dose (see footnote)	NRM \| NRS
141	2015	BWXT - Peterborough	Estimated public dose (see footnote)	NRM \| NRS
120	2016	BWXT - Peterborough	Estimated public dose (see footnote)	NRM \| NRS
99	2017	BWXT - Peterborough	Estimated public dose (see footnote)	NRM \| NRS
...	...	...	...	...
100	2017	SRBT	Tritium (HTO)	NRM \| NRS
78	2018	SRBT	Tritium (HTO)	NRM \| NRS
56	2019	SRBT	Tritium (HTO)	NRM \| NRS
34	2020	SRBT	Tritium (HTO)	NRM \| NRS
12	2021	SRBT	Tritium (HTO)	NRM \| NRS

	Year	Facility Name	Substance Name (English)	Units	Stack Emissions
9	2013	Nordion - Ottawa	Cobalt-60	GBq	0.0050
30	2014	Nordion - Ottawa	Cobalt-60	GBq	0.0050
51	2015	Nordion - Ottawa	Cobalt-60	GBq	0.0050
72	2016	Nordion - Ottawa	Cobalt-60	GBq	0.0060
93	2017	Nordion - Ottawa	Cobalt-60	GBq	0.0034

	Year	Facility Name	Substance Name (English)	Units	Stack Emissions
11	2013	Nordion - Ottawa	Iodine-125	GBq	0.2300
32	2014	Nordion - Ottawa	Iodine-125	GBq	0.1400
53	2015	Nordion - Ottawa	Iodine-125	GBq	0.1200
74	2016	Nordion - Ottawa	Iodine-125	GBq	0.2100
95	2017	Nordion - Ottawa	Iodine-125	GBq	0.0012

	Year \| Année	NPRI ID \| ID INRP	Company Name \| Raison Sociale	Facility Name \| Nom de l'installation	City \| Ville	CSD \| SDR	CA or CMA \| AR ou RMR	Economic Region \| Région économique	Province \| Province	Latitude \| Latitude	Longitude \| Longitude	Substance Name (English) \| Nom de substance (Anglais)	Substance Name (French) \| Nom de substance (Français)	Units \| Unités	Stack Emissions \| Émissions de cheminées	Direct Discharge \| Évacuations directes	Footnotes \| Notes de bas de page
50	2018	NaN	Cameco	Cameco Fuel Manufacturing	Port Hope	NaN	NaN	NaN	ON	43.954	-78.2748	Uranium	Uranium	kg	1.25	NRM \| NRS	NaN
72	2017	NaN	Cameco	Cameco Fuel Manufacturing	Port Hope	NaN	NaN	NaN	ON	43.954	-78.2748	Uranium	Uranium	kg	0.57	NRM \| NRS	NaN
50	2018	NaN	Cameco	Cameco Fuel Manufacturing	Port Hope	NaN	NaN	NaN	ON	43.954	-78.2748	Uranium	Uranium	kg	1.26	NRM \| NRS	NaN
72	2017	NaN	Cameco	Cameco Fuel Manufacturing	Port Hope	NaN	NaN	NaN	ON	43.954	-78.2748	Uranium	Uranium	kg	0.58	NRM \| NRS	NaN

	Year \| Année	NPRI ID \| ID INRP	Company Name \| Raison Sociale	Facility Name \| Nom de l'installation	City \| Ville	CSD \| SDR	CA or CMA \| AR ou RMR	Economic Region \| Région économique	Province \| Province	Latitude \| Latitude	Longitude \| Longitude	Substance Name (English) \| Nom de substance (Anglais)	Substance Name (French) \| Nom de substance (Français)	Units \| Unités	Stack Emissions \| Émissions de cheminées	Direct Discharge \| Évacuations directes	Footnotes \| Notes de bas de page
0	2021	3147.0	Canadian Nuclear Laboratories / Laboratoires N...	Chalk River Laboratories	Chalk River	Deep River	Renfrew	Kingston--Pembroke	ON	46.0554	-77.3628	Elemental Tritium (HT)	Tritium élémentaire	Bq	2.08E+12	NRM \| NRS	NaN
1	2021	3147.0	Canadian Nuclear Laboratories / Laboratoires N...	Chalk River Laboratories	Chalk River	Deep River	Renfrew	Kingston--Pembroke	ON	46.0554	-77.3628	Tritium (HTO)	Tritium (Eau tritiée)	Bq	2.49E+13	1.50E+13	NaN
2	2021	3147.0	Canadian Nuclear Laboratories / Laboratoires N...	Chalk River Laboratories	Chalk River	Deep River	Renfrew	Kingston--Pembroke	ON	46.0554	-77.3628	Carbon-14	Carbone-14	Bq	0.00E+00	NRM \| NRS	NaN
3	2021	3147.0	Canadian Nuclear Laboratories / Laboratoires N...	Chalk River Laboratories	Chalk River	Deep River	Renfrew	Kingston--Pembroke	ON	46.0554	-77.3628	Total noble gases	Total des gaz nobles	Bq-MeV	0.00E+00	NRM \| NRS	NaN
4	2021	3147.0	Canadian Nuclear Laboratories / Laboratoires N...	Chalk River Laboratories	Chalk River	Deep River	Renfrew	Kingston--Pembroke	ON	46.0554	-77.3628	Iodine-125	Iode-125	Bq	1.76E+06	NRM \| NRS	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
283	2013	NaN	Canadian Nuclear Laboratories / Laboratoires N...	Douglas Point	Tiverton	Kincardine	NaN	Stratford--Bruce Peninsula	ON	44.3267	-81.6000	Tritium (HTO)	Tritium (Eau tritiée)	Bq	1.59E+11	8.73E+10	NaN
284	2013	NaN	Canadian Nuclear Laboratories / Laboratoires N...	Douglas Point	Tiverton	Kincardine	NaN	Stratford--Bruce Peninsula	ON	44.3267	-81.6000	Particulate gross beta	Particules bêta brutes	Bq	NRM \| NRS	5.31E+07	NaN
285	2013	NaN	Canadian Nuclear Laboratories / Laboratoires N...	Douglas Point	Tiverton	Kincardine	NaN	Stratford--Bruce Peninsula	ON	44.3267	-81.6000	Estimated public dose (see footnote)	Dose estimée au public (voir note de bas de page)	mSv/a	0.0013	NRM \| NRS	Includes the entire Bruce site. Estimated publ...
286	2013	NaN	Canadian Nuclear Laboratories / Laboratoires N...	Nuclear Power Demonstration	Rolphton	NaN	NaN	NaN	ON	46.1868	-77.6578	Tritium (HTO)	Tritium (Eau tritiée)	Bq	6.86E+10	1.41E+11	NaN
287	2013	NaN	Canadian Nuclear Laboratories / Laboratoires N...	Nuclear Power Demonstration	Rolphton	NaN	NaN	NaN	ON	46.1868	-77.6578	Particulate gross beta	Particules bêta brutes	Bq	6.63E+04	9.76E+05	NaN

	Year	Facility Name	Substance Name (English)	Direct Discharge
268	2013	Chalk River Laboratories	Argon-41	NRM \| NRS
238	2014	Chalk River Laboratories	Argon-41	NRM \| NRS
208	2015	Chalk River Laboratories	Argon-41	NRM \| NRS
177	2016	Chalk River Laboratories	Argon-41	NRM \| NRS
141	2017	Chalk River Laboratories	Argon-41	NRM \| NRS
...	...	...	...	...
214	2015	Whiteshell Laboratories	Tritium (HTO)	NRM \| NRS
183	2016	Whiteshell Laboratories	Tritium (HTO)	NRM \| NRS
146	2017	Whiteshell Laboratories	Tritium (HTO)	NRM \| NRS
109	2018	Whiteshell Laboratories	Tritium (HTO)	NRM \| NRS
11	2021	Whiteshell Laboratories	Tritium (HTO)	NRM \| NRS

	Year	Facility Name	Substance Name (English)	Units	Direct Discharge	Footnotes
122	2018	Port Hope Project	Radium-226	Bq	7.000000e+05	NaN
124	2018	Port Hope Project	Radium-226	Bq	5.680000e+09	Releases from non-routine operations \| Rejets ...
159	2017	Port Hope Project	Radium-226	Bq	8.000000e+05	NaN
161	2017	Port Hope Project	Radium-226	Bq	1.590000e+10	Releases from non-routine operations \| Rejets ...
123	2018	Port Hope Project	Uranium	kg	5.000000e-01	NaN
125	2018	Port Hope Project	Uranium	kg	1.460000e+01	Releases from non-routine operations \| Rejets ...
160	2017	Port Hope Project	Uranium	kg	1.000000e-01	NaN
162	2017	Port Hope Project	Uranium	kg	1.101000e+02	Releases from non-routine operations \| Rejets ...

	Year	Facility Name	Substance Name (English)	Units	Direct Discharge
17	2013	Port Granby Project	Radium-226	Bq	5000000.0
19	2013	Port Hope Project	Radium-226	Bq	6200000.0
45	2014	Port Granby Project	Radium-226	Bq	5400000.0
48	2014	Port Hope Project	Radium-226	Bq	7700000.0
75	2015	Port Granby Project	Radium-226	Bq	4600000.0

	Year	Facility Name	Substance Name (English)	Units	Direct Discharge
8	2013	Chalk River Laboratories	Strontium-90	Bq	1.510000e+10
24	2013	Whiteshell Laboratories	Strontium-90	Bq	6.970000e+07
34	2014	Chalk River Laboratories	Strontium-90	Bq	2.260000e+11
54	2014	Whiteshell Laboratories	Strontium-90	Bq	6.610000e+07
64	2015	Chalk River Laboratories	Strontium-90	Bq	6.700000e+10

	Year	Facility Name	Substance Name (English)	Units	Direct Discharge
141	2017	Whiteshell Laboratories	Americium-241	Bq	5100000.0
176	2018	Whiteshell Laboratories	Americium-241	Bq	4210000.0
210	2019	Whiteshell Laboratories	Americium-241	Bq	20100000.0
244	2020	Whiteshell Laboratories	Americium-241	Bq	18000000.0

	Year	Facility Name	Substance Name (English)	Units	Direct Discharge
21	2013	Whiteshell Laboratories	Cesium-137	Bq	64000000.0
50	2014	Whiteshell Laboratories	Cesium-137	Bq	26600000.0
80	2015	Whiteshell Laboratories	Cesium-137	Bq	16500000.0
111	2016	Whiteshell Laboratories	Cesium-137	Bq	12800000.0
142	2017	Whiteshell Laboratories	Cesium-137	Bq	18900000.0

	Year	Facility Name	Substance Name (English)	Units	Direct Discharge
146	2017	Whiteshell Laboratories	Plutonium-238	Bq	8690000.0
181	2018	Whiteshell Laboratories	Plutonium-238	Bq	18400000.0
215	2019	Whiteshell Laboratories	Plutonium-238	Bq	48600000.0
249	2020	Whiteshell Laboratories	Plutonium-238	Bq	23900000.0

	Year	Facility Name	Substance Name (English)	Units	Direct Discharge
147	2017	Whiteshell Laboratories	Plutonium-239/240	Bq	12000000.0
182	2018	Whiteshell Laboratories	Plutonium-239/240	Bq	23200000.0
216	2019	Whiteshell Laboratories	Plutonium-239/240	Bq	47000000.0
250	2020	Whiteshell Laboratories	Plutonium-239/240	Bq	39400000.0

	Year \| Année	NPRI ID \| ID INRP	Company Name \| Raison Sociale	Facility Name \| Nom de l'installation	City \| Ville	CSD \| SDR	CA or CMA \| AR ou RMR	Economic Region \| Région économique	Province \| Province	Latitude \| Latitude	Longitude \| Longitude	Substance Name (English) \| Nom de substance (Anglais)	Substance Name (French) \| Nom de substance (Français)	Units \| Unités	Stack Emissions \| Émissions de cheminées	Direct Discharge \| Évacuations directes	Footnotes \| Notes de bas de page	Unnamed: 17	Unnamed: 18
0	2021	1147	Cameco	Rabbit Lake	Saskatoon	NaN	NaN	NaN	SK	58.1978	-103.7136	Uranium	Uranium	kg	NRM \| NRS	68.9	NaN	NaN	NaN
1	2021	1147	Cameco	Rabbit Lake	Saskatoon	NaN	NaN	NaN	SK	58.1978	-103.7136	Thorium-230	Thorium-230	MBq	NRM \| NRS	57.5	NaN	NaN	NaN
2	2021	1147	Cameco	Rabbit Lake	Saskatoon	NaN	NaN	NaN	SK	58.1978	-103.7136	Radium-226	Radium-226	MBq	NRM \| NRS	22.6	NaN	NaN	NaN
3	2021	1147	Cameco	Rabbit Lake	Saskatoon	NaN	NaN	NaN	SK	58.1978	-103.7136	Lead-210	Plomb-210	MBq	NRM \| NRS	76.6	NaN	NaN	NaN
4	2021	1147	Cameco	Rabbit Lake	Saskatoon	NaN	NaN	NaN	SK	58.1978	-103.7136	Polonium-210	Polonium-210	MBq	NRM \| NRS	43.1	NaN	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
220	2013	4866	Orano	McClean Lake	Saskatoon	NaN	NaN	NaN	SK	58.2611	-103.8025	Uranium	Uranium	kg	NRM \| NRS	1.8	NaN	NaN	NaN
221	2013	4866	Orano	McClean Lake	Saskatoon	NaN	NaN	NaN	SK	58.2611	-103.8025	Thorium-230	Thorium-230	MBq	NRM \| NRS	19.6	NaN	NaN	NaN
222	2013	4866	Orano	McClean Lake	Saskatoon	NaN	NaN	NaN	SK	58.2611	-103.8025	Radium-226	Radium-226	MBq	NRM \| NRS	6.0	NaN	NaN	NaN
223	2013	4866	Orano	McClean Lake	Saskatoon	NaN	NaN	NaN	SK	58.2611	-103.8025	Lead-210	Plomb-210	MBq	NRM \| NRS	74.4	NaN	NaN	NaN
224	2013	4866	Orano	McClean Lake	Saskatoon	NaN	NaN	NaN	SK	58.2611	-103.8025	Polonium-210	Polonium-210	MBq	NRM \| NRS	17.7	NaN	NaN	NaN

	Year	NPRI ID	Company Name	Facility Name	City	CSD	CA or CMA	Economic Region	Province	Latitude	Longitude	Substance Name (English)	Substance Name (French) \| Nom de substance (Français)	Units	Stack Emissions	Direct Discharge	Footnotes	Unnamed: 17	Unnamed: 18
15	2021	19397	Cameco	Cigar Lake	Saskatoon	NaN	NaN	NaN	SK	58.0686	-104.5406	Uranium	Uranium	kg	NRM \| NRS	0.0	NaN	NaN	NaN
203	2013	1147	Cameco	Rabbit Lake	Saskatoon	NaN	NaN	NaN	SK	58.1978	-103.7136	Lead-210	Plomb-210	MBq	NRM \| NRS	DL	Loadings were less than the detection limit \| ...	NaN	NaN

Radionuclide Release Analysis:¶

Table of Contents

Nuclear Power Plants¶

First Look at the Dataframe¶

Creating Geographic Reference Table¶

Cleaning of the Dataframe¶

Plotting by Substance¶

Estimated public dose¶

Tritium (HTO)¶

Carbon-14¶

Iodine-131¶

Particulate (gross beta/gamma)¶

Particulate gross alpha¶

Elemental Tritium (HT)¶

Total noble gases¶

Individual Plotting by Substance & Facility¶

Comparison with 2020 Dataframe¶

Nuclear Processing Facilities:¶

First Look at the Dataframe¶

Creating Geographic Reference Table¶

Cleaning of the Dataframe¶

Plotting by Substance¶

Estimated public dose¶

Uranium¶

Radium-226¶

Elemental Tritium (HT)¶

Tritium (HTO)¶

Cobalt-60¶

Iodine-125¶

Iodine-125¶

Xenon-133¶

Xenon-135¶

Xenon-135m¶

Individual Plotting by Substance & Facility¶

Comparison with 2020 Dataframe¶

Canadian Nuclear Laboratories¶

First Look at the Dataframe¶

Creating Geographic Reference Table¶

Cleaning of the Dataframe¶

Plotting by Substance¶

Carbon-14¶

Estimated public dose¶

Particulate gross alpha¶

Particulate gross beta¶

Radium-226¶

Strontium-90¶

Tritium (HTO)¶

Uranium¶

Americium-241¶

Argon-41¶

Cesium-137¶

Elemental Tritium (HT)¶

Iodine-125¶

Iodine-131¶

Plutonium-238¶

Plutonium-239/240¶

Total noble gases¶

Xenon-133¶

Individual Plotting by Substance & Facility¶

Comparison with 2020 Dataframe¶

Uranium Mines and Mills¶

First Look at the Dataframe¶

Creating Geographic Reference Table¶

Cleaning of the Dataframe¶

Plotting by Substance¶

Uranium¶

Thorium-230¶

Radium-226¶

Lead-210¶

Polonium-210¶

Individual Plotting by Substance & Facility¶

Comparison with 2020 Dataframe¶